I previously asked a question (or two) on the Works Discussion Forum relating to jLH Server and whether it would include server side scripts. The answers I got say that jLH Server is implemented from the same technology as existing NPP and NT Service and that server scripts would not be directly supported.

I'm currently looking at incorporating web functionality into some parts of our software. I see this can be done using oicgi with OpenInsight.

My question is this though: If one copy of OpenInsight (or later jLH) operates as a Web Server servicing many clients on an Intranet / Internet, is it able to timeslice through the requests it receives from the potentially many users? Currently in a Lan environment, each workstation is accessing the data directly, so each is processing essentially independently of the other, so no real performance or queuing issue. If a single copy of OI (or jLH) sits on the server and is accessed via oicgi or whatever, won't the requests get queued and processed sequentially? This would be a problem if one user requested a lengthy sort / select process to extract their report, followed closely by others just wanting to perform simple enquiries. Not a real issue in the Lan environment, but a real problem with several users gaining sequential access to the data via one copy of OpenInsight using oicgi - unless of course OI can now support multiple queues and can timeslice between them!!

Regards,

Matthew Jones.

[email protected]

Matt,

Let me check further into this for you. I am tempted to say that it would stack requests in a queue, since on the OI-side of things, OICGI logs in as a normal "user", however it might be possible to either log OICGI in under more than one account or use another workaround.

-Scott Revelation

View this thread on the Works forum...